Published Paper Details:

This paper presents the design and implementation of a smart spirometer for real-time respiratory health monitoring using Internet of Things (IoT) and machine learning techniques. The system employs a differential pressure-based sensing mechanism to measure airflow and estimate key lung function parameters such as Forced Expiratory Volume in one second (FEV1) and Forced Vital Capacity (FVC). An ESP32 microcontroller is used for data acquisition, processing, and wireless transmission of results to a mobile application for remote monitoring. To ensure reliable measurements, both hardware-level filtering and software-based signal processing methods are incorporated to reduce noise and improve stability. The calculated spirometric parameters are further utilized in a supervised machine learning framework, where clinically referenced threshold-based labeling is applied to generate structured training data. The trained model classifies respiratory conditions into normal and abnormal categories based on learned patterns. The proposed system provides a portable, low-cost, and user-friendly solution for continuous respiratory assessment and preliminary screening of conditions such as COPD and asthma. While the classification is guided by clinically established thresholds, the system demonstrates the integration of IoT and machine learning for accessible healthcare monitoring.